Re: [PATCH v2 06/23] media: adv748x: csi2: Implement get_fwnode_pad

[Steve Longerbeam <slongerbeam@gmail.com>]

Hi Jacopo,

On 11/28/19 3:53 AM, Jacopo Mondi wrote:
> Hi Steve,
>
> On Sun, Nov 24, 2019 at 11:06:46AM -0800, Steve Longerbeam wrote:
>> If the given endpoint fwnode passed to the .get_fwnode_pad() op is
>> the adv748x-csi2 TXA/TXB source endpoint, return the associated media
>> pad index ADV748X_CSI2_SOURCE. The adv748x-csi2 has no other media pads
>> that are associated with fwnode endpoints.
>>
>> Signed-off-by: Steve Longerbeam <slongerbeam@gmail.com>
>> ---
>>   drivers/media/i2c/adv748x/adv748x-csi2.c | 21 +++++++++++++++++++++
>>   1 file changed, 21 insertions(+)
>>
>> diff --git a/drivers/media/i2c/adv748x/adv748x-csi2.c b/drivers/media/i2c/adv748x/adv748x-csi2.c
>> index 2091cda50935..810085a1f2f0 100644
>> --- a/drivers/media/i2c/adv748x/adv748x-csi2.c
>> +++ b/drivers/media/i2c/adv748x/adv748x-csi2.c
>> @@ -228,6 +228,24 @@ static const struct v4l2_subdev_ops adv748x_csi2_ops = {
>>   	.pad = &adv748x_csi2_pad_ops,
>>   };
>>
>> +/* -----------------------------------------------------------------------------
>> + * media_entity_operations
>> + */
>> +
>> +static int adv748x_csi2_get_fwnode_pad(struct media_entity *entity,
>> +				       struct fwnode_endpoint *endpoint)
>> +{
>> +	struct v4l2_subdev *sd = media_entity_to_v4l2_subdev(entity);
>> +	struct adv748x_csi2 *tx = adv748x_sd_to_csi2(sd);
>> +
>> +	return endpoint->local_fwnode == tx->sd.fwnode ?
>> +		ADV748X_CSI2_SOURCE : -ENXIO;
> Couldn't you check if the endpoint port is either 10 or 11, as those
> are the only port numbers that provide a CSI-2 source pad ?
>
> In that case you could drop extending get_fwnode_pad() with th entity
> argument, as it is only used here (this one is actually the first user
> in the whole code base of this operation)

I don't think that's a very good idea. Entities that implement 
.get_fwnode_pad() in the future will likely need access to themselves in 
order to do the work. This implementation is a good example of that and 
is a better approach to checking the port number, because it is actually 
verifying that the endpoint fwnode object is owned by this device.

Steve

>
>> +}
>> +
>> +static const struct media_entity_operations adv748x_csi2_entity_ops = {
>> +	.get_fwnode_pad = adv748x_csi2_get_fwnode_pad,
>> +};
>> +
>>   /* -----------------------------------------------------------------------------
>>    * Subdev module and controls
>>    */
>> @@ -295,6 +313,9 @@ int adv748x_csi2_init(struct adv748x_state *state, struct adv748x_csi2 *tx)
>>   	/* Register internal ops for incremental subdev registration */
>>   	tx->sd.internal_ops = &adv748x_csi2_internal_ops;
>>
>> +	/* Register media_entity ops */
>> +	tx->sd.entity.ops = &adv748x_csi2_entity_ops;
>> +
>>   	tx->pads[ADV748X_CSI2_SINK].flags = MEDIA_PAD_FL_SINK;
>>   	tx->pads[ADV748X_CSI2_SOURCE].flags = MEDIA_PAD_FL_SOURCE;
>>
>> --
>> 2.17.1
>>